Source code of "Robust Transductive Few-shot Learning via Joint Message Passing and Prototype-based Soft-label Propagation".
- Download pretrained.zip and unzip it to get the pretrained folder, put it in the root directory, and put it on the same level as other folders
- Modify the script folder name of
exp_scripts
inmain()
in main.py and select the appropriate script list - Right-click pycharm to run
main.py
or executepython3 main.py
- The experimental results are saved in
/output/{exp_name}/{log_name}.log
The first experiment needs to rebuild the task according to the data set, and the built task is saved in /pretrained/preload
@article{wang2023robust,
title={Robust Transductive Few-shot Learning via Joint Message Passing and Prototype-based Soft-label Propagation},
author={Wang, Jiahui and Xu, Qin and Jiang, Bo and Luo, Bin},
journal={arXiv preprint arXiv:2311.17096},
year={2023}
}